Young adult smoking is a public health concern that has been largely understudied, particularly given that rates of smoking are highest in this age group (CADHS, 2008). Among young adults, light smoking (<5 cigs a day or <daily) is more common than heavy or daily smoking (Johnston et al., 2008). Recent studies have provided evidence for a link between light smoking and heavy drinking, with young adult nondaily smokers not only drinking more than daily smokers on days when drinking and smoking co-occurred (Harrison, et al, 2008), but also nondaily smokers were more likely to have had a recent binge episode (Harrison et al., 2009). However, the mechanisms linking these behaviors have not been investigated. Past research has shown that social factors are implicated in smoking initiation and progression and are thought to be a substantial influence on substance use behaviors in adolescents and young adults (e.g., Levinson et al., 2007). The proposed study will investigate the link between smoking, alcohol use, and social factors among young adults with the broader goal of informing prevention and intervention efforts with this population. Data for the current study will be drawn from four existing longitudinal studies that examine the initiation, progression, and cessation of substance use behaviors among college-attending young adults (N=613). Additionally, original cross-sectional data will be collected via an online survey of non-college attending young adults (N=500). First, to test a model of the hypothesized reciprocal role of alcohol use and smoking, thought to be mediated by interpersonal factors, age, and gender, cross-lagged path analysis will be employed, including both latent and manifest variables. Second, an instrument will be developed in order to measure the extent to which smoking is expected to facilitate social interactions. Extensive literature supports the role of expectancies (i.e., anticipated consequences of substance use) on future substance use behaviors. However, no multi-item instruments exist that measure this particular expectancy domain for young adults. Third, to test a prospective model of smoking and drinking progression parallel process latent growth curve analyses will be utilized. This aim will be addressed by examining the progression of smoking and drinking, their relationship to each other, and the mediational effects of interpersonal factors (exposure to smoking, significant others'smoking status, identity as a "social smoker"), intrapersonal factors (impulsivity, Nicotine Dependence, negative affectivity), and expectancies for substance use (utilizing the new instrument, a measure of alcohol use expectancies, and smoking and drinking expectancies). Finally, a model of the effect of interpersonal factors, intrapersonal factors, and alcohol use upon concurrent smoking will be tested with both college-attending and community young adults. The findings from the proposed study will serve to further our understanding of the relationship between smoking, drinking, and social factors and will serve to inform interventions that target both behaviors. PUBLIC HEALTH RELEVANCE: The proposed research will evaluate the relationship between alcohol use, social factors, and smoking among young adults, a population at risk for substance use and dependence. The findings will inform intervention and prevention efforts to address these dangerous health behaviors, which may yield considerable benefits in reducing the health burden of tobacco addiction and hazardous drinking.